1. Technical Field of the Invention
The present invention relates to the sending and receiving of electrical power data between different Energy Management Systems (EMS). Currently, there are about a dozen large EMS vendors and many smaller vendors. Each EMS vendor data format is inherently incompatible with any other EMS data format. More particularly, the present invention relates to an automated system and method to integrate, convert and maintain power grid information from various EMS vendors data formats into a single electrically connected database.
2. Description of the Prior Art
The electrical power generation industry has long enjoyed a regulated monopoly over the generation and transmission of electrical power and can influence service, policy and pricing to the consumers within their specific geographical area. Electrical power transfers are based on detailed network models encompassing only a specific geographical area with significantly less detailed network models of the surrounding area. Transferring power outside of the limited geographical area containing detailed electrical network models is problematical simply because of the lack of a unified network model. Consequently, such electrical power transfers are very limited in their range and frequency.
As a result of recent changes in Federal and State regulations, electric utilities are transitioning to a competitive market. The impact of competition is driving the market to seek out low cost energy producers and create a larger commodity market for electric power. Consequently, the Federal Energy Regulatory Commission (FERC) has mandated change through the issuance of Orders 888 and 889 (dated Apr. 24, 1996) and Order 2000 (dated March 2000) to encourage wholesale competition. Order 888 requires utilities to provide all power producers open access to the Transmission Grid and Order 889 requires utilities to establish electronic systems to share information about available transmission capacity. Order 2000 requires the creation of a Market-Reliability System to be implemented to ensure the reliability of the grid and provide non-discriminatory transmission access by 2005.
Accurately calculating the capacity of the transmission system and to efficiently manage congestion on the system requires that much larger detailed models of the network must be constructed that spans entire regions rather than one utility area. These models are the foundation upon which grid facility status and changes are monitored, risks managed and power transfers scheduled and routed. Without this fundamental model information (which does not exist), grid reliability and security is at risk and dependable market trading is impaired. Keeping models up to date will become an even more demanding task when the NERC policy on exchange of real-time network data is deployed on Sep. 1, 2001. Compliance with this policy will require automated model updating processes.
The operation (generation, transmission and distribution) of electrical power generation facilities employ a proprietary Energy Management System (EMS) that is tightly coupled to a particular operating system, computer manufacturer and a specific version of the associated application program. Examples of EMS vendors include the following: GE/Harris, ESCA, Siemens, Open Systems International (OSI), TeleGyr and Power Technologies (PSS/O). The nature of the various EMS vendors is to closely guard their proprietary secrets of operation, which includes the format of their data. The present state of the art is such that sending or receiving electrical power from different EMS vendors is difficult at best to perform without significant advance notice. Minimal transactions between different EMS formats are typical due the time required to translate the data formats from one EMS vendor""s :format to another.
The lack of a detailed power system network topology when spanning three or more EMS vendor data formats can result in electrical power scheduling failures. Consider the network topology for these three bordering companies: company A, company B, and company C. Company A has topological connections only to company B. Company B has topological connections only to company C. When company A schedules a power transfer to company C, company B does not know about the power transfer although the actual wires are in the company B area. The scheduling error occurs when company B attempts to send power through the same lines that are already in use by companies A and C. This case is rare, however, it has been known to occur and is the direct result of not having a single model that encompasses all three of the companies areas.
Generally, at least 24 hours advance notice are required by any given EMS vendor to send or receive electrical power to a different EMS due to the problems associated with dynamically balancing the generation capacity and demand load for a given time interval starting from different data formats. This time period can be significant in the case of an unpredictable emergency caused by natural phenomena such as a hurricane or unusually high temperatures during the summer months.
The problem of sharing data between different EMS vendors in a timely fashion has been partly addressed in the development of a standard exchange format known as the Common Information Model (CIM) sponsored by the Electrical Power Research Institute (EPRI). The CIM is an abstract model that represents all of the major objects typically contained in an EMS information model. A typical EMS data format will contain power system resources such as: companies, divisions, substations, transformers, generators, busbars, A/C lines, D/C Lines, capacitor banks, reactor banks, energy consumers (loads), breakers, switches, conductors, connectivity nodes, fuses, jumpers and grounds at various voltage levels.
The current CIM was based on U.S. Pat. No. 5,604,982 to Nuttall et al. Feb. 18, 1997. This patent relies on a single source information model that separates the data format and relationships through the use of surrogate keys as described on page 11. Changing a surrogate key would make all objects associated with the key change. The primary concern was to allow various programs access to a single data source. The primary concern here is with network models and building a single network model database from raw data of various formats. There is a lack of instruction on how to convert various data formats into the CIM format and further a lack of instruction on how to merge data from multiple CIM compliant databases. Nuttall et al. does teach how to construct a discrete hierarchical relationship based on physical quantification. However, these relationships have inherent limitations that must be expanded in order to produce and maintain composite network models.
Significant effort is required to convert the various EMS data formats into the CIM due in part because of the different terminology used to describe the same physical equipment. A database capable of recalling specific power systems equipment requires significant programming effort simply because of the vast amount of equipment involved and the numerous types of equipment to be classified. The various EMS""s differ in their internal data format and the relationships between the various elements or equipment. The actual migration of raw EMS extracted data into the CIM format requires computer programming skills and specific electrical power modeling knowledge.
When an EMS vendor converts their internal data format to the CIM format, the geographically bordering EMS must also convert their data to the CIM format. Electrical power transfers are generally limited to about 1000 miles due to the degradation of electrical flow though the wires. This distance could span multiple EMS data formats. Typically, power transfers occur only on bordering EMS""s and rarely three or more different EMS vendor data formats are involved in any given electrical power flow transaction.
The data contained in the CIM format from two different EMS data types must be integrated into a single power grid model to properly analyze a power transfer request. The process of integrating two or more power grid databases is not included in the CIM specification. This currently requires extensive manual effort to insure that the electrical connections are correct and properly electrically terminated. None of the current EMS vendors have converted their proprietary power systems data to the CIM format.
One of the primary manual tasks involved in the power grid consolidation is resolving the naming of substations and the various power systems equipment types such that the names are unique. For example, most EMS network topology models in any given area have a substation named xe2x80x9cairportxe2x80x9d. A given EMS format may require that all substation names are unique and typically the foreign substation is renamed. Yet another instance is when power utility companies merge with bordering companies having the same or different EMS data formats. Again, unique naming of the power systems equipment presents the problem even when the newly acquired utility has the same EMS format. Another example of naming convention problems is when one EMS vendor allows duplicate names for specific power systems equipment and this is an error condition in the foreign EMS vendors"" system. Additionally, the character length of the names for the various power systems equipment varies from vendor to vendor. An EMS vendor type may only allow eight characters for the equipment and another EMS vendor may allow up to 16 characters or more for the same equipment. Consolidating the 16 or more character names into an eight-character field presents additional naming problems. The naming conventions, given the number of substations and related power system resources, are of great concern when these number in the thousands and tens of thousands.
Tie Lines are inter-substation transmission lines that typically connect between two different EMS data formats. These are used to transmit or receive electrical power between different companies. Transformers are similarly used between companies when the source and target voltage levels are different. The target locations are considered outside of the source substation""s network. Modeling is generally sparse in terms of the power system equipment detail for the target network topology due to the limited information provided by the target company and the proprietary nature of the EMS in the target substation.
Another manual process involves extracting only a portion of the foreign network topology due to company ownership boundaries. This is labor intensive again due to the requirement that the resulting network topology must have proper connectivity. One or more alternating or direct current Tie Lines connects between the source and target area substations. The target area substation also has connections to other substations in the target company and perhaps other companies such that severing the connections during a merging process causes non-terminated nodes that must be resolved. Typically, several manual iterations of this process are required to properly electrically terminate the target substation""s externally linked equipment. This problem is further compounded by the modeling efforts of the source substation""s sparse representation of the target substation""s network. Integrating the source and target substations requires that the source substation must first remove the sparse representation of the target substation before adding the more detailed representation of the target substation to obtain the complete network topology.
The actual electrical power grids are represented in a wide variety of data formats. The planning and network topology models show increasing levels of power grid detail. EMS vendors do not provide the same level of detail for all of the power systems related equipment. The manual effort required to generate a fully populated network topology is monumental when there are no starting references other than the application programs controlling the power flow.
The graphic representation of a substation is referred to as a One-Line. The One-Line diagram contains industry standard symbols that represent specific devices commonly found in a substation. These manually or automatically generated diagrams are currently used for source material when modeling external power systems equipment. Another method currently used is to manually text edit one or more specific EMS raw data file(s) from the target substation(s) into a corresponding text formatted file to join the data. These manual processes are again duplicated when new power systems equipment is added or removed to update the current representation of the physical hardware. Obviously, these methods are time consuming and prone to human error. Indeed, this virtual xe2x80x9cTower of Babylonxe2x80x9d represented in the various EMS formats is a major contributing factor for the government intervention.
Based on the aforementioned problems described above, the technical problem exists to:
1) Find an automated method to integrate power system data from multiple EMS vendor""s data formats into a single power systems database that is properly electrically connected, easily updateable and maintains EMS vendor specific data format confidentiality;
2) Find a method to automatically convert from one EMS vendor""s data format to another EMS vendor""s data format; and
3) Facilitates compliance with the Federal Energy Regulatory Commission (FERC) orders 888, 889 and 2000.
The present invention has been reduced to practice for several EMS vendor""s data formats using programming languages from Microsoft Visual C++ and Visual Basic. Database operations use Microsoft Access format and optionally any Open Database Connectivity (ODBC) or Data Access Objects (DAO) may be used. Other platforms and development software may also be used. The present invention is directed to a process which:
(a) Imports Energy Management System vendor specific raw data and converts the data into a Common Information Model (CIM) database;
(b) Adds various connectivity extensions to the CIM database;
(c) Produces an electrically connected CIM power systems database;
(d) Automatically integrates two or more CIM power systems databases;
(e) Detects and facilitates user resolution of possibly disconnected devices encountered when integrating the power grid databases;
(f) Checks equipment electrical connectivity and distinctively enumerates invalid connections;
(g) Facilitates updating the CIM power grid database through a graphical user interface to display and edit power systems equipment; and
(h) Exports a CIM database into a different EMS vendor data format.
The present invention has several advantages over the prior art of manually integrating one line diagrams or the text based editing of raw data. One advantage is in the reduction or elimination of human errors introduced in the process of text editing or manually developing the data using One-Line diagrams.
Another advantage is the reduction or elimination of power scheduling errors caused by the lack of a single network model that encompasses the entire area in which the scheduled power transfers occur.
Another advantage of the present invention is the speed of the automated process that converts the various EMS data formats into a Common Information Model. The computer processing time required in converting a vendor specific EMS raw data set into the CIM data format is significantly less than manual methods once the vender specific EMS conversion xe2x80x9cfilterxe2x80x9d to the CIM has been developed.
Yet another advantage of the present invention is the ease of updating of the power systems equipment representation using a graphical user interface with various hierarchical views capable of downward tree expansion.
Yet another advantage of the present invention is the resolution of unique naming convention problems encountered when integrating foreign electrical network topologies through automatic cross-reference generation of inter-substation Tie Lines and related power systems equipment.
Yet another advantage of the present invention is the automatic conversion of one EMS vendors power system data format into another EMS vendors format while maintaining confidentiality of EMS specific data formats.
Still another advantage of the present invention is the automated integration of multiple EMS vendors"" power systems databases into a single electrically connected database.
It is therefore an object of the present invention to provide a method and apparatus that inputs the raw data from any EMS data format and automatically converts it into the CIM format with specific integration extensions.
It is another object of the invention to provide a method and apparatus to automatically convert the CIM""s power systems data format to another EMS vendor""s power systems data format.
It is another object of the invention to provide a method and apparatus to allow the updating of the power systems equipment in the database for maintenance and simulation studies.
It is yet another object of the present invention to provide a method and apparatus to automatically integrate two power systems databases into a single electrically connected database. This process can be infinitely repeated such that all of the electrical power can be contained in a single database containing the entire network topology and associated CIM data.
It is yet another object of the present invention to provide a method and apparatus to detect and provide correction for disconnected devices as a result of integrating two power systems databases.
It is still another object of the invention to provide a method of documenting the physical power systems equipment. Because the original data is extracted from a running EMS application, the results of the imported data can be used to visually verify the in-house representation of the physical power system equipment present in the EMS system.
The foregoing disclosure and the content of the drawings are merely illustrative of the principals of this invention and are not to be interpreted in a limiting sense. Accordingly, the CIM is subdivided into several packages with the current focus on the class relationships found in the Core, Topology, and Wires packages. One skilled in the art would find that all of the remaining packages and future CIM extensions may be similarly processed. Additionally, non-proprietary data formats from other industries conforming to common information model architecture could be similarly processed. Still further objects and advantages will become apparent from a consideration of the ensuing description of a preferred embodiment of the invention and accompanying drawings wherein like reference numbers represent like parts of the invention. The present invention is best understood when correlating the table extensions as follows:
1) A xe2x80x9cCIMxe2x80x9d prefix indicates that it is the xe2x80x9cBasexe2x80x9d or work table containing additional information and that extraneous information is removed when the table is loaded in to the table of the same name, and
2) A xe2x80x9cPsyxe2x80x9d prefix indicates that the table was created and was not directly derived from the CIM specification.